Smooth muscle cell FTO regulates contractile function

Abstract

The fat mass and obesity gene (FTO) is a N⁶-methyladenosine RNA demethylase that was initially linked by Genome-wide association studies to increased rates of obesity. Subsequent studies have revealed multiple mass-independent effects of the gene, including cardiac myocyte contractility. We created a mouse with a conditional and inducible smooth muscle cell deletion of Fto (Myh11 Cre⁺ Fto^fl/fl) and did not observe any changes in mouse body mass or mitochondrial metabolism. However, the mice had significantly decreased blood pressure (hypotensive), despite increased heart rate and sodium, and significantly increased plasma renin. Remarkably, the third-order mesenteric arteries from these mice had almost no myogenic tone or capacity to constrict to smooth muscle depolarization or phenylephrine. Microarray analysis from Fto^−/−-isolated smooth muscle cells demonstrated a significant decrease in serum response factor (Srf) and the downstream effectors Acta2, Myocd, and Tagln; this was confirmed in cultured human coronary arteries with FTO siRNA. We conclude Fto is an important component to the contractility of smooth muscle cells.

NEW & NOTEWORTHY We show a key role for the fat mass obesity (FTO) gene in regulating smooth muscle contractility, possibly by methylation of serum response factor (Srf).

INTRODUCTION

Fat mass and obesity gene (FTO) is an N⁶-methyladenosine RNA demethylase (1, 2) and was first identified from GWAS studies where single nucleotide polymorphisms within the locus were associated with increased risk of obesity (3, 4). Additional human and murine studies appear to support this initial finding (4–6), although there are additional studies that implicate several genes in body mass regulation (7). The Fto locus has also been implicated in osteoarthritis (8, 9) and Alzheimer’s (10, 11), and renal disease (12), indicating that body mass regulation may indeed be a secondary function of the gene. Fto polymorphisms and deletion have been shown to impact the prevalence of cardiovascular disease (3, 13–22).

Recently, our laboratory has demonstrated that Fto deletion from endothelial cells (ECs) protected mice from metabolic disease after a 12-wk high-fat diet (23). The protection included lipocalin-type prostaglandin-D synthase (L-PGDS) regulation and subsequent prostaglandin-D₂ (PGD₂) signaling. It was discovered that decreased PGD₂, due to elevated FTO expression in obesity, reduced AKT signaling and insulin sensitivity. The diminished PGD₂ signaling also decreased myogenic tone and increased blood pressure. Importantly, there was no overt phenotype when Fto was deleted from ECs until obesity was induced through high-fat-diet feeding.

Here we demonstrate that Fto deletion from smooth muscle cells (SMCs) is associated with reduced resistance artery myogenic tone and blood pressure independent from changes in body mass. We hypothesize this is due to a significant decrease in the ability of SMCs to constrict.

MATERIALS AND METHODS

Mice

Fto^fl/fl mice were a generous gift from Dr. Jens Bruning (Max Planck Institute, Germany; 24) and crossbred with a tamoxifen-inducible myosin heavy chain Cre (Myh11 creER^T2), herein referred to as Myh11 Cre (25). Only male mice were included in the study because of the position of the Myh11 Cre on the Y-chromosome (25). This Cre (25) does present some issues, including the inability to test the effects of Fto genetic deletion on females and the possibility of the Cre altering the intestinal SMCs, variables that newly developed SMC Cres will be able to test once available (26).

For our experiments, littermates were used in all instances. At 6 wk of age, Myh11 Cre⁺ Fto^fl/fl mice were injected daily with vehicle (peanut oil; PO) or 1 mg of tamoxifen (T) for 10 consecutive days; 6-wk-old Myh11 Cre⁻ Fto^fl/fl mice were also injected with 1 mg of tamoxifen for 10 consecutive days. Fto^+/+ and Fto^−/− mice were used as described in Ref. 5. All mice had a 12-h:12-h light/dark cycle, were kept at 22–24°C, and had access to water and food ad libitum. At 20 wk of age, mice were euthanized by CO₂ inhalation and used for experiments following the guidelines of the University of Virginia Institutional Animal Care and Use Committee (IACUC). All animal-related procedures and experiments used in this study were included in an animal protocol that was submitted and approved by the University of Virginia IACUC.

Immunofluorescence

Immunofluorescence procedures and protocols are previously outlined (23). In brief, a primary mouse monoclonal antibody to Fto (abcam 92821 1:200) was used to analyze the presence or absence of Fto in paraffin cross sections of third-order mesenteric arteries. Primary vascular SMCs were derived from thoracic aorta of global Fto^+/+ and Fto^−/− mice (n = 3/genotype, pooled), fixed, and stained for filamentous actin (via phalloidin; Thermo A12380; 1:1,000), Acta2 (Sigma A2547: 1:200), and cell nuclei using DAPI (Invitrogen, P36931). Blocking solution contained 0.5% BSA and 5% goat serum in PBS, and antibody solution contained 0.5% BSA, 5% goat serum, and 0.25% Triton-X-100. Blocking was performed for 1 h at room temperature, primary antibody incubation was performed overnight at 4°C, and secondary antibodies were added for 1 h at room temperature. Fluorescent images were quantified using arbitrary fluorescent unit analysis on FIJI as previously described (27). All images are representative z-stacks obtained with an Olympus Fluoview 1000.

Body Weight, Length, and Epidydimal Fat

Weekly body weight measurements were made from 7 to 20 wk of age for all mice to analyze weight gain over time. At 20 wk, one final body weight recording was taken, as well as body length (nose to anus) measurements and epidydimal fat pad weights postisolation.

siRNA-Mediated FTO Knockdown In Vitro

Primary human coronary artery smooth muscle cells (HCSMCs) were purchased from Cell Applications (350 K-05a). All cells were maintained under standard cell culture conditions (5% CO₂ at 37°C) in smooth muscle cell base media (Lonza; CC-3181) and supplemented with growth factors (Lonza; CC-4149) and 5% fetal bovine serum. All cells used for experiments were below 10 passages. To induce a contractile state before siRNA application, HCSMCs were starved of serum for 24 h before transfection, and 48 h during the transfection process in the presence of 1 nM dexamethasone (72 h total). To knockdown Fto, 1 nmol/L scrambled or FTO siRNA (Origene, SR324951) was used with lipofectamine (RNAiMAX, Invitrogen) according to the manufacturer’s protocol in nonantibiotic, nonserum-supplemented media that included growth factors. Cells were harvested 48 h after transfections.

Seahorse Assay

Mitochondrial function of HCSMCs treated with scrambled siRNA and FTO siRNA (stated earlier) was quantified using a Seahorse XF24 Flux Analyzer (Agilent Technologies). Protocols and procedures for extracellular flux analysis were previously described (23).

Quantitative PCR

Total RNA isolation was performed using a Zymo Quick-RNA miniprep kit (11-327). Isolated RNA (500 ng) was reverse transcribed to cDNA using superscript reverse transcriptase (Thermofisher 18090050). Taqman probes (all from Thermofisher) information are as follows: B2M (Hs0018742), B2m (Mm00437762_m1), FTO (Hs01057140_m1), Fto (Mm00488755_m1), SRF (Hs01065257_g1), Srf (Mm00491032_m1), MYOCD (Hs00538076_m1), Myocd (Mm00455051_m1), TAGLN (Hs06633192_s1), Tagln (Mm00441661_g1), KLF4 (Hs00358836), ACTA2 (Hs00426835_g1), and Acta2 (Mm00725412_s1). Transcript expression was analyzed using the ΔΔCt method normalized to B2M.

Radiotelemetry

Radiotelemetry procedures were performed as previously described (23) for 20-wk-old Myh11 Cre⁺ Fto^fl/fl (+Tmx), Myh11 Cre⁻ Fto^fl/fl (+Tmx), and Myh11 Cre⁺ Fto^fl/fl (+PO) mice.

Myogenic Tone

Third-order mesenteric arteries, cleaned of all fat, were taken from 20-wk-old Myh11 Cre⁺ Fto^fl/fl (+Tmx), Myh11 Cre⁻ Fto^fl/fl (+Tmx), and Myh11 Cre⁺ Fto^fl/fl (+PO) mice and were subject to pressure steps with (passive) and without (active) Ca²⁺-free buffer as previously described (23). NS309 dilations, phenylephrine, and KCl constrictions were performed after myogenic tone.

RNA Microarray

Bioinformatical upstream analysis of microarray data from thoracic aortae of global Fto^+/+ and Fto^−/− mice were used as previously published (23) and publicly available through Gene Expression Omnibus (GEO series accession No. GSE138840). These data were used to perform ingenuity pathway analysis (IPA) to identify possible Fto-affected transcripts such as the transcription factor serum response factor (SRF) and further downstream targets.

Serum Analysis

Blood was extracted via cardiac puncture, and serum was analyzed for sodium, renin, and blood urea nitrogen (BUN) levels as previously described (28).

Western Blot Analysis

Human coronary artery cells (Cell Applications, 350 K-05a) were lysed in cold RIPA buffer, consisting of 50 mmol/L Tris·HCL, 150 mmol/L NaCl, 5 mmol/L EDTA, 1% deoxycholate, and 1% Triton-X-100 in PBS with pH adjusted to 7.4 and containing protease inhibitor cocktail (Sigma) and P2/P3 phosphatase inhibitor cocktail (Sigma). Cell lysate was subjected to dounce homogenization and sonicated with three pulses lasting 2 s each and subsequently incubated at 4°C for 1 h to solubilize proteins. Samples were centrifuged for 10 min at 12,000 g to pellet cell debris. Protein concentration was determined via bicinchoninic acid (BCA) assay (Pierce), and 20 µg of protein was loaded into each lane of an SDS gel electrophoresis system using 8% Bis-Tris gels (Invitrogen). Protein was transferred to the nitrocellulose membrane for immunoblotting. Membranes were blocked for 1 h at room temperature with 3% BSA in Tris-buffered saline and then incubated overnight in 5% BSA in Tris-buffered saline with primary antibodies, human anti-myosin light chains (Sigma M4401 1:500), and human phospho-Myl9 (ser19; Thermofischer PA5-17726 1:200). Membranes were washed and incubated in LiCOR IR dye secondary antibodies for 1 h and both visualized and quantified using the LiCOR odyssey with Image Studio software. Protein abundance was normalized to total protein staining (revert total protein stain LiCOR).

Statistics

Statistical analyses were performed using Graphpad Prism v.9.3.1. For continuous variables, normality and homogeneity of variance were assessed by Shapiro–Wilk. After the confirmation of homogeneous variances and normality, two-group comparisons for means were performed by two-sided Student’s t test, and multigroup comparisons for means were performed by two-way analysis of variance (ANOVA) with Holm-Sidak multiple comparison test, respectively. For data that did not pass either normality or equal variance tests, multigroup comparisons were performed by Friedmann’s two-way ANOVA on ranks test with Tukey’s post hoc test. Data are represented as means ± SE with a P value of <0.05 used as the threshold for statistical significance.

RESULTS

Using the same Fto^fl/fl mice we had previously used for endothelial cell (EC) deletion (23), we bred Fto^fl/fl mice to Myh11 Cre⁺ mice for inducible, specific deletion of FTO from SMCs (25, 27, 29). We previously demonstrated the expression of FTO in SMCs by immunocytochemistry and in situ hybridization (23). FTO was no longer evident in the Myh11 Cre⁺ Fto^fl/fl (+Tmx) mice by immunocytochemistry and significantly decreased Fto mRNA (Fig. 1A). Controls in all experiments were the Myh11 Cre⁻ Fto^fl/fl (+Tmx) and Myh11 Cre⁺ Fto^fl/fl (PO) mice to control for the effects of tamoxifen and genotype, respectively. In all cases, mice were injected with Tmx or PO at 6 wk for 10 consecutive days, and all experiments were performed on 20-wk-old mice (Fig. 1B). Because previous work in humans had correlated SNPs in FTO with increased BMI (30), we measured body mass and body length, which were unchanged across groups (Fig. 1, C and D). Weight gain over 12 wk, as well as fat pad mass, were measured with no changes observed between genotypes (data not shown). Finally, metabolic analysis from humans with SNPs within the FTO locus implicated altered cellular metabolism (31). For this reason, we used HCSMCs and siRNA to knockdown FTO. Extracellular flux assays did not reveal any differences in overall mitochondrial function when Fto was absent in these cells (Fig. 1E). Thus, initial genotyping of mice lacking FTO in SMCs, or human cultured SMCs with suppression of FTO, did not identify an overt phenotype that had been previously associated with FTO.

Figure 1.

Smooth muscle cell FTO decreases blood pressure but does not affect body mass or cellular metabolism. A: representative immunofluorescence image of smooth muscle FTO (magenta) expression in third-order mesenteric arteries of Myh11 Cre⁺ Fto^fl/fl (+PO, top) but not in Myh11 Cre⁺ Fto^fl/fl (+Tmx, bottom). Images (n = 4, 4, 4) were quantified using arbitrary fluorescence units per SMC nuclei (DAPI; cyan); scale bar represents 50 µm, (***P = 0.0008 Cre⁺ vs. Cre⁺ and ***P = 0.0009 Cre⁺ vs. Cre⁻). Mesenteric arteries (endothelium and smooth muscle both present) were used for mRNA expression of Fto (*P = 0.041 Cre⁺ vs. Cre⁺ and *P = 0.05 Cre⁺ vs. Cre⁻). B: experimental timeline showing 10 consecutive day of tamoxifen injections, with tissue harvest occurring at 20 wk. C: body mass at 20 wk was analyzed with no significant differences seen between groups (n = 19, 7, 8). D: body length was measured at 20 wk and revealed no FTO dependent body length effects (n = 19, 7, 8). Data represent means ± SE for statistical analysis two-way ANOVA with Tukey’s post hoc test was performed; P < 0.05 was defined as significant. E: analysis of mitochondrial function using extracellular flux analysis (n = 4) showed no significant differences between scrambled vs. FTO siRNA at any measurement point. FTO knockdown was confirmed via qPCR and normalized to B2M, (**P = 0.0047). Radiotelemetric analysis of inactive (F) and active (G) blood pressure with systolic and diastolic average pressures. H: mean arterial pressure (MAP). n = 3, 4, 5, respectively, for F–I. For F, systolic, *P = 0.048 and diastolic, *P = 0.049; for G, systolic, *P = 0.050 and *P = 0.049, respectively, with diastolic, *P = 0.021 and **P = 0.0094, respectively; for H, *P = 0.041. Blood plasma was used in J–L (n = 5 for each condition) to examine sodium (J), renin (K), and blood urea nitrogen (BUN; L). For active renin in K, *P = 0.031 and *P = 0.029, respectively. For F–L, data represent means ± SE for statistical analysis two-way ANOVA with Tukey’s post hoc test was performed; *P < 0.05 was defined as significant. FTO, fat mass and obesity gene; PO, peanut oil; SMC, smooth muscle cell.

Because of the key role for SMCs in maintaining peripheral resistance, we examined blood pressure during both active and inactive periods in Myh11 Cre⁺ Fto^fl/fl (+Tmx) mice using radiotelemetry. A significant reduction in blood pressure was observed during inactive and active periods (Fig. 1, F–H). A slight elevation in heart rate and plasma sodium was also observed (Fig. 1, I and J) with a significant increase in plasma renin (Fig. 1K). Thus, the mice were incapable of maintaining homeostatic blood pressure despite numerous compensatory mechanisms. Kidney function as assessed by BUN was unchanged (Fig. 1L).

To determine if the spontaneous hypotension was due to the inability of the SMCs to contract, we used pressure myography on third-order mesenteric arteries. Third-order mesenteric arteries are well-described resistance arteries that greatly contribute to the peripheral resistance component of blood pressure. The myogenic tone of the arteries from Myh11 Cre⁺ Fto^fl/fl (+Tmx) arteries was dramatically decreased, with no observable contractile response to increases in pressure, which was evident in the active tone component (Fig. 2A). The passive tone was unaltered, indicating the extracellular matrix was not affected, adding to the conclusion that the loss of myogenic tone response is due to SMC contractility (Fig. 2A). In addition to myogenic tone on third mesenteric arteries, we observed in mice lacking Fto in smooth muscles cells an inability for mesenteric arteries or carotid arteries to constriction in response to depolarization with KCl (Fig. 2, B and C) or smooth muscle-specific agonist (phenylephrine; Fig. 2, D and E). Dilation to the endothelium-specific agonist NS309 was unchanged (Fig. 2F).

Figure 2.

Blood pressure and vascular tone can be regulated by smooth muscle cell FTO. Experiments were performed on 20-wk-old mice. Colors for bars and lines: Myh11 Cre⁻ Fto^fl/fl (+Tmx), blue; Myh11 Cre⁺ Fto^fl/fl (+PO), purple; and Myh11 Cre⁺ Fto^fl/fl (+Tmx), orange. A: pressure myography analysis in third-order mesenteric arteries myogenic tone analysis [*P < 0.05 for pressures 80–120 mmHg for Myh11 Cre⁺ Fto^fl/fl (Tmx) compared with both controls; n = 5, 6, 5 in each group, respectively] with active [*P < 0.05 for pressures 80–120 mmHg for Myh11 Cre⁺ Fto^fl/fl (Tmx) compared with both controls] and passive pressures. Vascular reactivity assays using third-order mesenteric arteries (B, D, and F) or carotids (C and E) to test (B and C) 30 mM KCl constriction (B, *P = 0.025 with n = 3, 4, 3; C, *P = 0.035 and *P = 0.015 respectively, with n = 4, 3, 4); 10⁻⁶ phenylephrine constriction (D, *P = 0.026 and *P = 0.011, respectively, with n = 3, 3, 4; E, *P = 0.019 and *P = 0.023, respectively, with n = 3, 3, 4); or NS309 (F) dilation (no significant differences between groups; n = 3, 5, 7). G: RNA microarray analysis reveals a decrease in SRF a known regulator of smooth muscle cell phenotypes. H: immunofluorescence staining of vascular smooth muscle cells from the thoracic aorta isolated from global Fto^−/− vs. Fto^+/+ mice stained with DAPI (blue), F actin (yellow), and ACTA2 (magenta). Scale bar denotes 10 µm (n = 4). qPCR analysis [n = 4 for each condition from either Fto^−/−/Fto^+/+ mouse aortas or HCSMCs transfected with Fto siRNA (siFTO)/control siRNA (siCntrl)] of downstream targets of Srf/SRF (mice, *P = 0.011; cells, *P = 0.0172; I) show decreases in Myocd/MYOCD (mice, *P = 0.024; cells, *P = 0.022; J), Tagln/TAGLN (mice, *P = 0.031; cells, *P = 0.027; K), and Acta2/ACTA2 (mice, ***P = 0.0009; cells, ***P = 0.0007; L) expression. There was no statistical change in KLF4 (M). In N, full-length Western blot analysis of HCSMCs transfected with Fto siRNA (siFTO) or control siRNA (siCntrl) are shown probed for myosin light chain (MLC) or phosphorylated MLC (pMLC). Total protein is also shown with molecular weight markers noted in filled black arrowhead and MLC denoted in the open arrowhead. Blots are quantified in O. For ratio of pMLC to MLC, *P = 0.046. In N and O, n = 4 for each condition. Data in I–M and O represent means ± SE for statistical analysis unpaired, two-sided Student’s t test was performed. FTO, fat mass and obesity gene; HCSMCs, human coronary artery smooth muscle cells; PO, peanut oil.

We next used aortas isolated from Fto^−/− and Fto^+/+ mice and performed RNA microarray analysis to identify possible changes in gene expression relevant for contractile function [Fig. 2G; (23)]. One of the most notable decreases in gene expression revealed by ingenuity pathway analysis was for serum response factor (SRF), a master regulatory transcription factor for contractile proteins in SMCs. Freshly isolated SMCs from Fto^−/− mice confirmed this finding by demonstrating a perinuclear accumulation of ACTA2 compared with the stress fibers observed in SMCs derived from Fto^+/+ mice (Fig. 2H). mRNA from Fto^−/− aortas confirmed knockdown of Srf (Fig. 2I), Myocd (Fig. 2J), Tagln (Fig. 2K), and Aacta2 (Fig. 2L). To determine if this was the case in human smooth muscle cells, we again used HCSMCs and knocked down FTO with siRNA (Fig. 1E). Similar to the results from Fto^−/− mice, SRF, MYOCD, TAGLN, and ACTA2 mRNA were also significantly reduced (Fig. 2, I–L). mRNA levels of KLF4, an upstream modulator of SRF (32–34), were unchanged with loss of Fto (Fig. 2M). Finally, we measured protein levels of myosin light chain (MLC) and its phosphorylated form in HCSMCs with and without Fto (Fig. 2N). Although total MLC was decreased, total phosphorylation was not (Fig. 2O), demonstrating another compensatory mechanism by smooth muscle cells attempting to increase contraction when Fto is absent. Thus, mice lacking FTO in SMCs have a spontaneous hypotension that corresponds with the inability of the resistance arteries to constrict; this may be due to alterations in SRF.

DISCUSSION

Here we present a novel role for FTO as a regulator of SMC contractility. This was based on several observations both in vivo and in vitro: 1) Myh11 Cre⁺ Fto^fl/fl mice had a significant decrease in baseline systolic and diastolic pressures; 2) Myh11 Cre⁺ Fto^fl/fl mice had essentially no myogenic tone, i.e., inability to respond to increases in intraluminal pressure with constriction; 3) Myh11 Cre⁺ Fto^fl/fl lacked capacity to respond to KCl or agonist-induced constriction; 4) RNA microarray data from freshly isolated aortic Fto^−/− SMCs pointed to a significant downregulation of SRF, and finally 5) in vitro HCSMCs with FTO knockdown showed decreased gene expression of SRF, MYOCD, TAGLN, and ACTA2, canonical markers for contractile SMCs. Cumulatively, the data presented point toward a novel role for Fto as a possible upstream regulator of SRF and SMC-regulated arterial control of contraction and blood pressure. As such, Fto may influence cell differentiation between contractile and synthetic vascular SMC phenotypes, as it has been demonstrated to be important in epithelial-to-mesenchymal cell transition (35).

Because Fto is an m6A RNA demethylase and has been shown to localize to the nucleus, it is possible SRF mRNA is a substrate of Fto (1, 36). Srf is considered a master regulator of SMC contractile phenotypes (37–40), a critical function of the vasculature. SRF transcription directly regulates the levels of α-smooth muscle actin, Sma22a, as well as other SMC-specific genes (41). Srf has also been shown to be regulated by histone lysine methylations that affect downstream myocardin factors and SMC differentiation (42, 43). Conversely, histone methylation by TET2 has been shown to inhibit SMC plasticity and promote SMC-specific gene expression. Most importantly, Fto-dependent mRNA methylation can regulate cardiac myocyte contraction (2), giving precedence for the conclusion that vascular SMC Fto could regulate contractility by demethylation of SRF mRNA. Further studies are needed to interrogate the specific Fto demethylation sites on SRF mRNA that directly affect RNA stability, protein expression, and cellular phenotype. Given the severe loss of contractility seen with loss of Fto, future studies are also needed to assess the effect that loss of Fto has on smooth muscle cell calcium signaling. Indeed, besides Srf, we cannot discount other reasons that contractility of SMCs may be lost. There are numerous other possibilities including posttranslational controls via protein phosphorylation, potassium channel activity, direct suppression of MYOCD expression, differential regulation of Srf binding, and changes in matrix composition. Indeed, McDonald et al. have shown that Srf-dependent gene expression is primarily regulated by epigenetic mechanisms that control Srf binding to CArG elements, not by regulating SRF levels (41, 44, 45).

Previous work performed by our laboratory showed EC Fto knockout had no effect on myogenic tone in normal chow diet-fed mice (23). However, research presented here shows a dramatic loss of both myogenic tone and KCl-induced constriction in Myh11 Cre⁺ Fto^fl/fl mice. The vascular phenotypes presented in this manuscript, bare resemblance to phenotypes previously noted in mice lacking Srf. These studies used the same inducible Myh11Cre, allowing for physiological comparisons between smooth muscle cell loss of Srf and the loss of Fto. Specifically, the loss of myogenic tone (23) and constrictive response to KCl (46) seen in Myh11 Cre⁺ Srf^fl/fl mice mimic the vascular reactivity data presented here. Myh11 Cre⁺ Srf^fl/fl mice show decreases in arterial blood pressure (46), a phenotype strikingly similar to Myh11 Cre⁺ Fto^fl/fl mice. Myh11 Cre⁺ Srf^fl/fl mice also show decreased mRNA and protein levels of ACTA2, a similar result to what we present in vitro with siFTO HCSMCs (46, 47). Given the similar phenotypes between mice lacking FTO and mice lacking SRF in SMCs, it is likely that these two genes reside in the same regulatory pathway involved in SMC contractility. Based on the previously published data and our novel data, we postulate here that Fto regulates SRF mRNA stability and abundance, therefore playing an early role in the regulation of SMC phenotypes.

These data are in line with FTO overexpression in murine cardiac myocytes having increased contractility (2), indicating a possible larger role for Fto in regulation of muscle contraction. Although we cannot completely discount decreased contractility of cardiac myocytes playing a role in the phenotype we describe, there are several reasons against this conclusion, including 1) Myh11 Cre has been extensively shown to be specific to SMCs, especially resistance artery SMCs; 2) pressure myography of isolated resistance arteries demonstrated essentially no active tone, where SMCs contract in response to pressure, and 3) mice lacking Fto in SMCs had an increased heart rate and serum renin, a typical response to low peripheral resistance. If Fto was altering cardiac myocyte contractility, the heart rate would not be able to compensate. For these reasons, we conclude the physiological outcome of FTO deletion in SMCs was specific to the contractility of these cells and provides evidence for a larger role for Fto regulating muscle contraction.

At the physiological level, this work identifies Fto as an important regulator of SMC contraction. Our data show Fto may control SRF expression, a master regulator of SMC contractile phenotypes. However, additional work demonstrating a direct effect on SRF mRNA demethylation is required, e.g., using epitranscriptomics. Based on the data presented herein, it is possible Fto could serve as a novel drug target to protect from hypertension and increased peripheral resistance seen in cardiovascular disease.

DATA AVAILABILITY

The data that support the findings of this study are available from the corresponding author upon request. RNA microarray data have been made publicly available through gene Expression and can be accessed using the GEO series accession No. GSE138840.

GRANTS

This work was supported by Deutsche Forschungsgemeinschaft Grant IRTG 1902 (to N.K. and A.G.) and National Heart, Lung, and Blood Institute Grants T32 HL007284 (to M.A.L., M.E.G., L.A.B., and V.S.) and HL088554 (to B.E.I.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

M.A.L., N.K., R.A.D., A.G., and B.E.I. conceived and designed research; M.A.L., N.K., M.E.G., L.A.B., V.S., A.S., and R.A.D. performed experiments; M.A.L., N.K., L.A.B., V.S., and R.A.D. analyzed data; M.A.L., N.K., N.L. A.G., and B.E.I. interpreted results of experiments; M.A.L. and N.K. prepared figures; M.A.L. and N.K. drafted manuscript; M.A.L., N.K., M.E.G., N.L., A.G., and B.E.I. edited and revised manuscript; M.A.L., N.K., M.E.G., L.A.B., V.S., A.S., R.A.D., N.L., A.G., and B.E.I. approved final version of manuscript.

APPENDIX

Figure A1 presents a listing of major resources pertaining to animals, antibodies, cultured cells, data and code availability, and other.

Figure A1.

Major resources detail animals, antibodies, cultured cells, data and code availability, and other.